TMS target spot positioning method and system based on function connection weight

Abstract

The invention provides a TMS target spot positioning method and system based on a function connection weight. The method comprises the following steps: preprocessing structural magnetic resonance image data and functional magnetic resonance image data of a plurality of subjects; in the preprocessed structural magnetic resonance image data, obtaining a region of interest corresponding to a TMS stimulation target of each subject to obtain a plurality of regions of interest; obtaining a functional connection matrix between the plurality of regions of interest and whole brain regions in the preprocessed functional magnetic resonance image data; obtaining a function connection weight matrix between the depression scale score data and the function connection matrix; mapping the function connection weight matrix into a function connection matrix to obtain an estimated curative effect map of the whole brain region; and carrying out clustering on voxels of a left DLPFC area in the estimated curative effect diagram, and determining a clustering center as a TMS treatment target. According to the method, the problem that the TMS used in the related technology is lack of individualized accurate stimulation target spots is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of transcranial magnetic stimulation target location, and in particular to a TMS target location method and system based on functional connection weights. Background Art

[0002] Transcranial Magnetic Stimulation (TMS) is a non-invasive neuromodulation technology that is used to study human neurophysiology and treat neurological diseases. TMS mainly performs magnetic stimulation on the left dorsal lateral prefrontal cortex (DLPFC). Related studies have shown that TMS can change the functional connectivity between the DLPFC and distal subcortical areas, thereby improving depression. At present, the TMS technology used in related technologies lacks individualized precise stimulation targets. Summary of the invention

[0003] The present invention provides a TMS target positioning method and system based on functional connection weights, which are used to solve the problem that TMS used in related technologies lacks individualized precise stimulation targets.

[0004] The present invention provides a TMS target location method based on functional connection weights, comprising: Obtaining depression scale score data, structural magnetic resonance imaging data, and functional magnetic resonance imaging data of several subjects; preprocessing the structural magnetic resonance imaging data and the functional magnetic resonance imaging data to obtain preprocessed structural magnetic resonance imaging data and preprocessed functional magnetic resonance imaging data; In the preprocessed structural magnetic resonance imaging data, a region of interest corresponding to the TMS stimulation target of each subject is obtained to obtain a plurality of regions of interest; Acquire a functional connection matrix between the plurality of regions of interest and the whole brain regions in the preprocessed functional magnetic resonance imaging data; Obtaining a functional connection weight matrix between the depression rating scale score data and the functional connection matrix, wherein the functional connection weight matrix is ​​used to characterize the degree of correlation between the functional connection matrix and the depression rating scale score data; The functional connection weight matrix is ​​mapped to the functional connection matrix to obtain an estimated therapeutic effect map of the whole brain area; the voxels in the left DLPFC area in the estimated therapeutic effect map are clustered, and the cluster center is determined as the TMS treatment target.

[0005] According to the TMS target location method based on functional connection weights provided by the present invention, the region of interest corresponding to the TMS stimulation target of each subject is obtained in the pre-processed structural magnetic resonance imaging data, and several regions of interest are obtained, including: Obtain the coil position and coil posture of each subject during TMS treatment; According to the coil position and the coil posture, electric field simulation is performed in the cerebral cortex of the preprocessed structural magnetic resonance image data to obtain a simulated electric field; Obtaining the point with the maximum simulated electric field value in the simulated electric field as the TMS stimulation target point; In the cerebral cortex, the region of interest corresponding to the TMS stimulation target of each subject is obtained to obtain a plurality of regions of interest.

[0006] According to the TMS target positioning method based on functional connection weights provided by the present invention, the step of obtaining a functional connection matrix between the plurality of regions of interest and the whole brain regions in the preprocessed functional magnetic resonance imaging data comprises: Extracting blood oxygen level dependent signals of the plurality of regions of interest, the plurality of whole brain regions and the plurality of whole brain voxels to obtain time series matrices of the plurality of regions of interest, the plurality of whole brain regions and the plurality of whole brain voxels; Obtaining the Pearson correlation coefficient between the time series matrices of the plurality of regions of interest and the time series matrices of the entire brain region to obtain a first functional connectivity matrix; Obtaining the Pearson correlation coefficient between the time series matrix of the whole brain region and the time series matrix of the whole brain voxels to obtain a second functional connectivity matrix; The functional connectivity matrix is ​​determined according to the first functional connectivity matrix and the second functional connectivity matrix.

[0007] According to the TMS target positioning method based on functional connection weights provided by the present invention, the step of obtaining a functional connection weight matrix between the depression rating scale score data and the functional connection matrix includes: Based on the partial least squares regression model, a functional connection weight matrix between the depression rating scale score data and the functional connection matrix is ​​obtained.

[0008] According to the TMS target location method based on functional connection weights provided by the present invention, the structural magnetic resonance imaging data and the functional magnetic resonance imaging data are preprocessed to obtain the preprocessed structural magnetic resonance imaging data and the preprocessed functional magnetic resonance imaging data, including: Acquiring brain tissue image data from the structural magnetic resonance image data; Acquiring the functional magnetic resonance imaging data after time correction and head motion correction to obtain corrected functional magnetic resonance imaging data; The brain tissue image data and the corrected functional magnetic resonance image data are registered on the same standard template to obtain the preprocessed structural magnetic resonance image data and the preprocessed functional magnetic resonance image data respectively.

[0009] The TMS target location method based on functional connection weights provided by the present invention further includes: The extracted blood oxygen level dependent signal is smoothed by a bandpass filter.

[0010] The present invention also provides a TMS target positioning system based on functional connection weights, comprising: An image acquisition module is used to acquire depression scale score data, structural magnetic resonance imaging data, and functional magnetic resonance imaging data of several subjects; preprocess the structural magnetic resonance imaging data and the functional magnetic resonance imaging data to obtain preprocessed structural magnetic resonance imaging data and preprocessed functional magnetic resonance imaging data; A region acquisition module, used to acquire a region of interest corresponding to a TMS stimulation target of each subject in the preprocessed structural magnetic resonance imaging data, and obtain a plurality of regions of interest; A first matrix acquisition module is used to acquire a functional connection matrix between the plurality of regions of interest and the whole brain regions in the preprocessed functional magnetic resonance imaging data; A second matrix acquisition module is used to acquire a functional connection weight matrix between the depression scale score data and the functional connection matrix, wherein the functional connection weight matrix is ​​used to characterize the degree of correlation between the functional connection matrix and the depression scale score data; The target acquisition module is used to map the functional connection weight matrix into the functional connection matrix to obtain an estimated therapeutic effect map of the whole brain area; cluster the voxels in the left DLPFC area in the estimated therapeutic effect map, and determine the cluster center as the TMS treatment target.

[0011] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the TMS target location method based on functional connection weights as described in any one of the above is implemented.

[0012] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described TMS target localization methods based on functional connection weights.

[0013] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method for TMS target localization based on functional connection weights as described above is implemented.

[0014] The TMS target location method and system based on functional connection weights provided by the present invention constructs a functional connection weight matrix that characterizes the degree of correlation between the functional connection matrix and the depression scale score data based on the functional connection data of the actual TMS stimulation target and the whole brain region before treatment and the improvement of the therapeutic effect after TMS treatment. And the functional connection weight map is used to predict potential effective treatment targets, realizing an accurate and individualized TMS treatment of depression target location method. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 This is one of the flow charts of the TMS target localization method based on functional connection weights provided by the present invention.

[0017] Figure 2 It is a schematic diagram of the principles of structural magnetic resonance imaging and functional magnetic resonance imaging provided by the present invention.

[0018] Figure 3 This is the second flow chart of the TMS target localization method based on functional connection weights provided by the present invention.

[0019] Figure 4 It is an estimated therapeutic effect diagram of the TMS target positioning method based on functional connection weights provided by the present invention.

[0020] Figure 5 It is a schematic diagram of the principle of the leave-one-out cross-validation method provided by the present invention.

[0021] Figure 6 It is a schematic diagram of the results of linear regression prediction based on PLSR scores provided by the present invention.

[0022] Figure 7 It is a schematic diagram comparing the therapeutic effects of the TMS target localization method provided by the present invention and the TMS target localization method in the related art.

[0023] Figure 8 It is a structural schematic diagram of the TMS target positioning system based on functional connection weights provided by the present invention.

[0024] Fig. 9 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the descriptions used in this way can be interchanged where appropriate, so that the embodiments can be implemented in a sequence other than that illustrated or described in the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or modules is not necessarily limited to those steps or modules that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. The naming or numbering of the steps that appear in the present application does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The process steps that have been named or numbered can change the execution order according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved. The division of modules in this application is a logical division. There may be other division methods when it is implemented in actual applications. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection between units can be electrical or other similar forms, which are not limited in this application. In addition, the modules or submodules described as separate components may or may not be physically separated, may or may not be physical units, or may be distributed in multiple circuit units, and some or all of the units may be selected according to actual needs to achieve the purpose of the present application.

[0027] In an embodiment of the present invention, each subject is a person who has received iTBS treatment for a period of time. iTBS treatment is Stanford therapy, which is a brain nerve regulation technology that adjusts brain neuron activity through transcranial magnetic stimulation (TMS) or other methods. Current studies have shown that Stanford therapy (iTBS) may have certain diagnostic and therapeutic effects on some mental and psychological diseases. In an embodiment of the present invention, the subject may be a person with certain symptoms of depression.

[0028] It should be noted that in the embodiments of the present invention, depression is only used as an exemplary illustration of a mental illness. Based on the concept of the present invention, the TMS target localization method based on functional connection weights can also be applied to the TMS target localization system for other mental illnesses.

[0029] Combine the following Figure 1-Figure 9 The specific contents of the present invention are described.

[0030] The TMS target location method based on functional connection weight provided by the present invention can be applied to the TMS target location system.

[0031] Figure 1 It is a flow chart of a TMS target localization method based on functional connection weights provided by the present invention, including steps S101-S105.

[0032] Step S101, obtaining depression scale score data, structural magnetic resonance imaging data and functional magnetic resonance imaging data of several subjects; preprocessing the structural magnetic resonance imaging data and the functional magnetic resonance imaging data to obtain preprocessed structural magnetic resonance imaging data and preprocessed functional magnetic resonance imaging data.

[0033] In an embodiment of the present invention, structural magnetic resonance imaging data and functional magnetic resonance imaging data of several subjects before TMS treatment are obtained; as well as the position and posture of the stimulation coil during the actual TMS treatment; and depression scale score data before treatment, one week after treatment, and two weeks after treatment.

[0034] Step S102, obtaining a region of interest corresponding to the TMS stimulation target of each subject in the preprocessed structural magnetic resonance image data, and obtaining a plurality of regions of interest.

[0035] Specifically, the TMS stimulation target is the target position of the subject when the subject received TMS treatment before. The region of interest is located in the pre-processed structural magnetic resonance imaging data and is composed of a region within a preset range around the TMS stimulation target.

[0036] Step S103, obtaining a functional connection matrix between a plurality of regions of interest and whole brain regions in the preprocessed functional magnetic resonance imaging data.

[0037] In an embodiment of the present invention, the functional connectivity matrix is ​​used to characterize the functional connectivity relationship between a plurality of regions of interest in the preprocessed structural magnetic resonance imaging data and whole brain regions in the preprocessed functional magnetic resonance imaging data.

[0038] Step S104, obtaining a functional connection weight matrix between the depression scale score data and the functional connection matrix, where the functional connection weight matrix is ​​used to characterize the degree of correlation between the functional connection matrix and the depression scale score data.

[0039] Step S105, mapping the functional connection weight matrix to the functional connection matrix to obtain an estimated therapeutic effect map of the whole brain area; clustering the voxels in the left DLPFC area in the estimated therapeutic effect map, and determining the cluster center as the TMS treatment target.

[0040] In a possible implementation, step S101, preprocessing the structural magnetic resonance imaging data and the functional magnetic resonance imaging data to obtain preprocessed structural magnetic resonance imaging data and preprocessed functional magnetic resonance imaging data, includes steps S101A-S101C.

[0041] Step S101A, acquiring brain tissue image data from the structural magnetic resonance image data.

[0042] Specifically, the preprocessing of the collected structural magnetic resonance image data includes: brain tissue image data extraction. Exemplarily, the skull and other non-brain tissues in the structural magnetic resonance image are removed using the BET brain extraction algorithm, thereby extracting brain tissue image data.

[0043] Step S101B, acquiring functional magnetic resonance imaging data after time correction and head motion correction.

[0044] Specifically, head motion correction: Since the acquisition time of functional magnetic resonance images usually lasts for a long time, the head movement of the subject during the acquisition process can easily cause the image to shift, so it is necessary to use a head motion correction algorithm to correct the functional magnetic resonance images at each time point, by estimating the motion parameters of each scanning time point, and then repositioning all functional magnetic resonance images to the same time reference point. Time correction: Since functional magnetic resonance images are scanned by layer, each layer is acquired at a different time, but in subsequent analysis it is generally believed that all layers are acquired at the same time, so time correction is required. After the above time correction and head motion correction, the corrected functional magnetic resonance image is obtained.

[0045] like Figure 2 FIG. 1 is a schematic diagram of the principle of structural magnetic resonance imaging and functional magnetic resonance imaging provided by an embodiment of the present invention. Figure 2The upper middle part is the functional magnetic resonance images collected at multiple time points. Exemplarily, the number of time points is set to 180 in the embodiment of the present invention.

[0046] Step S101C, registering the brain tissue image data and the corrected functional magnetic resonance image data to the same standard template, and obtaining the preprocessed structural magnetic resonance image data and the preprocessed functional magnetic resonance image data, respectively.

[0047] Specifically, in the process of configuring the brain tissue image data, the brain tissue image data is registered to a standard template or other templates. The registration process generally uses linear registration and nonlinear registration.

[0048] Specifically, during the configuration of the whole-brain imaging data, since the size and shape of the head and brain tissues of the subjects are different, in order to facilitate the subsequent statistical analysis between the subjects, it is necessary to align the whole-brain imaging data of all subjects to a standard spatial template. Here, the registration process file in the configuration process of the brain tissue imaging data is directly used to align the whole-brain imaging data to the standard template.

[0049] In a possible implementation, step S102, in the preprocessed structural magnetic resonance imaging data, obtaining a region of interest corresponding to the TMS stimulation target of each subject, and obtaining a plurality of regions of interest, including: obtaining the coil position and coil posture of each subject during TMS treatment; performing electric field simulation in the cerebral cortex of the preprocessed structural magnetic resonance imaging data according to the coil position and coil posture to obtain a simulated electric field; obtaining a point with a maximum simulated electric field value in the simulated electric field as a TMS stimulation target; obtaining a region of interest corresponding to the TMS stimulation target of each subject in the cerebral cortex, and obtaining a plurality of regions of interest.

[0050] Specifically, electric field simulation was performed according to the coil position and coil posture of each subject during actual TMS treatment.

[0051] Exemplarily, the position of the actual stimulation target on the cerebral cortex is determined. Since the stimulation coil is placed on the cerebral scalp, the position of the contact point between the coil plane and the cerebral scalp is collected. In order to further analyze the brain connection characteristics on the cerebral cortex, it is necessary to map the position and posture of the stimulation coil to the cerebral cortex. First, the gray matter of the brain is segmented and the cerebral cortex is reconstructed according to the structural magnetic resonance image, and then the electric field simulation is performed in the cerebral cortex of the preprocessed structural magnetic resonance image data according to the coil position and coil posture. In this way, a simulated electric field individualized for each subject can be obtained. Then, the point with the largest simulated electric field value is determined on the cerebral cortex as the center point of the stimulation target region of interest, and the radius of the region of interest is set to 10 mm. Since the calculation of the simulated electric field is only performed on the segmented cerebral cortex, it can be ensured that the region of interest corresponding to the TMS stimulation target is located on the cerebral cortex. The regions of interest obtained for several subjects can be found in Figure 2 The stimulation target region of interest in the middle and lower parts,

[0052] In a possible implementation, step S103, a functional connection matrix between a plurality of regions of interest and whole brain regions in the preprocessed functional magnetic resonance imaging data is obtained, such as Figure 3 As shown, steps S301-S304 are included.

[0053] Step S301, extracting blood oxygen level dependent signals of a number of regions of interest, whole brain regions and whole brain voxels to obtain time series matrices of the number of regions of interest, whole brain regions and whole brain voxels.

[0054] Exemplarily, blood-oxygen-level-dependent (BOLD) signals are extracted from several regions of interest (spheres with a radius of 10 mm centered on the TMS stimulation target), whole-brain regions (brain regions defined by the brain network group atlas), and whole-brain voxels (whole-brain voxels under the MNI template). The MNI template is the standard template in the above-mentioned embodiment of the present invention.

[0055] In the embodiment of the present invention, the number of subjects N is 35, the number of whole brain regions is 246, the size of the whole brain voxel is 2mm, the MNI template has 235375 voxels, and the functional magnetic resonance imaging scans 180 time points. Therefore, the time series matrix sizes of several regions of interest, whole brain regions, and whole brain voxels are 35×180, 246×180, and 235375×180, respectively.

[0056] In a possible implementation, step S103 further includes: smoothing and filtering the extracted blood oxygen level dependent signal through a bandpass filter.

[0057] In the embodiment of the present invention, since the activity frequency of the brain is generally around 0.01Hz-0.1Hz, a bandpass filter is used to smooth the BOLD signal during the BOLD signal extraction process, with a low-pass filter frequency of 0.01Hz, a high-pass filter frequency of 0.1Hz, and a time interval of 2s. At the same time, spatial smoothing with a half-peak full width of 6 mm is also performed to reduce spatial noise and increase the signal-to-noise ratio.

[0058] Step S302, obtaining the Pearson correlation coefficient between the time series matrices of the several regions of interest and the time series matrices of the whole brain regions, and obtaining a first functional connectivity matrix.

[0059] Step S303, obtaining the Pearson correlation coefficient between the time series matrix of the whole brain region and the time series matrix of the whole brain voxels to obtain a second functional connectivity matrix.

[0060] Step S304: determining a functional connectivity matrix according to the first functional connectivity matrix and the second functional connectivity matrix.

[0061] In steps S303 and S304, the Pearson correlation coefficients between the time series matrices of several regions of interest and the time series matrices of the whole brain regions are calculated to obtain a first functional connectivity matrix with a size of 35 × 246. The Pearson correlation coefficients between the time series matrices of the whole brain regions and the time series matrices of the whole brain voxels are calculated to obtain a second functional connectivity matrix with a size of 246 × 235375.

[0062] In an embodiment of the present invention, in order to make the functional connectivity matrix more visual, the Pearson correlation coefficient in the functional connectivity matrix is ​​subjected to Fisher-z transformation to obtain a functional connectivity image.

[0063] In a possible implementation, step S104, obtaining a functional connection weight matrix between depression scale score data and the functional connection matrix, includes: obtaining a functional connection weight matrix between depression scale score data and the functional connection matrix based on a partial least squares regression model.

[0064] Based on the partial least squares regression model, i.e., the PLSR model, the correlation between TMS stimulation targets and functional connectivity improvement was explored to obtain the weight of each brain region in the brain network group map in the correlation analysis, i.e., the functional connectivity weight matrix. Among them, the functional connectivity improvement is represented by the above functional connectivity matrix.

[0065] The PLSR model can be expressed by the following two formulas:

[0066] in and The matrices represent the first functional connectivity matrix and the second functional connectivity matrix, respectively. and They are and The partial least squares estimate of the matrix, and is the corresponding residual matrix. and Each of them is represented by a score matrix ( and ) and a loading matrix ( and ) and multiply them to get, and It can be considered and The representation in the projected low-dimensional space. Representation Matrix The amount of active ingredients, i.e. .

[0067] Assumptions and They are and The first principal component of , The CCA model focuses on the correlation between the independent variables and the dependent variables after projection, and its optimization goal is , while PCR focuses on the method interpretability of the data before and after projection, so its optimization goal is , PLSR combines these two methods, optimizing the correlation between the independent variable and the dependent variable after projection, and also focusing on retaining the data information before and after projection. Its optimization goal is .

[0068] The optimization objective of PLSR can be expressed as:

[0069] It is easy to solve is a symmetric matrix The eigenvector corresponding to the maximum eigenvalue of is a symmetric matrix After solving the first pair of related principal components, we can then and The principal components are regressed and modeled respectively, and the following are obtained:

[0070] And according to and The correlation between and The regression model is:

[0071] Then the least squares method is used to solve the three equations to obtain the load factors corresponding to the principal component scores, and we get:

[0072] Then the residuals that cannot be explained by this pair of principal components in the regression process are and As new and , and then continue to perform regression to solve new principal components and loading factors until the residual Achieve a certain accuracy requirement or have the number of principal components reach the number of effective components .

[0073] Solve all After the principal components are and It is expressed as:

[0074] So, we got about The regression equation for and Principal components in projected space.

[0075] Will and The principal components in the projected space were determined as the functional connectivity weight matrix between the depression scale score data and the functional connectivity matrix.

[0076] Dimensions after PLSR reduces the data dimension , that is, the number of principal components is an important parameter that determines the prediction performance of the model. In order to select the optimal number of components, the PLSR model with different numbers of components was trained 1000 times, and its regression index was calculated to obtain the number of components corresponding to the model with the best regression performance. In order to prevent the PLSR model from overfitting on the training data, a permutation test was also performed on the model.

[0077] Step S105, mapping the functional connection weight matrix to the functional connection matrix to obtain an estimated therapeutic effect map of the whole brain area; clustering the voxels in the left DLPFC area in the estimated therapeutic effect map, and determining the cluster center as the TMS treatment target.

[0078] In step S105, the mapping relationship of the functional connection weight matrix is ​​applied to the functional connection matrix. The matrix is ​​reduced in dimension to obtain Then use the linear regression model to predict the potential therapeutic effect of each voxel, and finally restore the one-dimensional vector to a three-dimensional brain image to obtain Figure 4 The estimated therapeutic effect map is shown. By clustering the voxels belonging to the left DLPFC region in the therapeutic effect map, the cluster center is determined as the individualized optimal treatment target, that is, the TMS treatment target.

[0079] In an embodiment of the present invention, for step S105, the estimated therapeutic effect map of the whole brain region is obtained according to the functional connection weight moment. It is necessary to first verify the predictability of the PLSR model for the therapeutic effect of depression, and use a linear regression model to predict the therapeutic effect after treatment based on the PLSR score after dimensionality reduction.

[0080] For each model evaluation of PLSR and linear regression, the leave-one-out cross-validation method was used for evaluation (e.g. Figure 5 As shown in Figure 1. Each time, one subject is selected and not used for training. The remaining N-1 (N is 35) subjects are used to train the linear regression model. Then the trained model is used to predict the subjects that were not used for training. This training and prediction is performed for each subject to obtain the prediction results for all subjects. Then the R2 score of the predicted results of all subjects and the actual results of all subjects is calculated as the model evaluation result of the leave-one-out validation.

[0081] The PLSR model was used to explore the correlation between the functional connectivity of the actual stimulation target and the brain area of ​​the brain network group map and the efficacy of depression, and the corresponding principal component weights were obtained. , i.e., the functional connection weight matrix, which indicates the contribution of each brain region in the brain network group map to predict the potential therapeutic effect of depression. The closer the weight is to 1, the stronger the functional connection between the target stimulation point and the brain region, the better the corresponding therapeutic effect will be. Through this functional connection weight matrix, the functional connection matrix between the brain regions and brain region voxels of N subjects can be reduced in dimension, that is, the 35 The functional connectivity matrix is ​​mapped into 35 The vectors are then combined into The training data of is linearly regressed with the corresponding 35 subjects’ depression evaluation efficacy to obtain a linear regression model. The specific regression process can be expressed as:

[0082] For each voxel, , find the best parameters and Then for the new known but Unknown data can be directly predicted Possible values .

[0083] The linear regression prediction results based on PLSR scores in the embodiments of the present invention are described below.

[0084] The linear regression prediction results based on PLSR scores are as follows Figure 6 As shown, the evaluation process adopts the leave-one-out method. Figure 6 The left side shows the prediction result of the linear regression model trained using the depression scale score reduction rate after one week of treatment as label data. Figure 6 On the right, the depression scale score reduction rate after two weeks of treatment is used as the label data. It is found that the model trained using the depression scale score reduction rate after two weeks of treatment has a significant positive correlation in its prediction results (two weeks: r=0.474, p=0.004*). However, the model trained using the depression scale score reduction rate after one week of treatment has a high P value and a low statistical effect size (one week: r=0.292, p=0.089).

[0085] The correlation between the distance between the optimal treatment target and the actual stimulation target and the reduction rate of the depression rating scale was used to evaluate the effectiveness of the TMS target positioning method proposed in the present invention. Figure 7 As shown in the figure, the horizontal axis is the Euclidean distance between the optimal treatment target and the actual stimulation target, and the vertical axis is the TMS efficacy. After two weeks of treatment, there was a significant negative correlation between the distance between the targets and the reduction rate of the depression rating scale (r=-0.343, p=0.044, Figure 7 The lower right side shows that the closer the actual stimulation target is to the optimal treatment target, the better the corresponding therapeutic effect will be, which verifies the effectiveness of the treatment target located by the method of the present invention. However, no significant negative correlation was found in the depression scale score reduction rate after one week of treatment (r=-0.234, p=0.176), which is consistent with the result that the PLSR score has a weak predictive performance for the depression scale score reduction rate after one week of treatment. In addition, it was compared with the traditional positioning method based on the functional connectivity of the sgACC brain region. Figure 7 The results on the left show that the treatment target localization method based on the functional connectivity of the sgACC brain region did not show a significant negative correlation.

[0086] The results show that the TMS target localization method based on functional connection weights proposed in the present invention is better than the localization method based on sgACC brain region functional connection in the related art.

[0087] By adopting the above method, the present invention can achieve at least one of the following beneficial effects: based on the functional connection data between the actual TMS stimulation target and the whole brain area before treatment, and the improvement of the therapeutic effect after TMS treatment, a functional connection weight matrix representing the correlation between the functional connection matrix and the depression scale score data is constructed. And the functional connection weight map is used to predict potential effective treatment targets, realizing an accurate and individualized TMS treatment depression target positioning method.

[0088] The TMS target localization system based on functional connection weights provided by the present invention is described below. The TMS target localization system based on functional connection weights described below and the TMS target localization method based on functional connection weights described above can be referenced to each other.

[0089] like Figure 8 As shown, the present invention also provides a TMS target positioning system based on functional connection weights, comprising: The image acquisition module 810 is used to obtain depression scale score data, structural magnetic resonance imaging data and functional magnetic resonance imaging data of several subjects; pre-process the structural magnetic resonance imaging data and functional magnetic resonance imaging data to obtain pre-processed structural magnetic resonance imaging data and pre-processed functional magnetic resonance imaging data.

[0090] The region acquisition module 820 is used to acquire the region of interest corresponding to the TMS stimulation target of each subject in the pre-processed structural magnetic resonance image data, and obtain a plurality of regions of interest.

[0091] The first matrix acquisition module 830 is used to acquire the functional connection matrix between several regions of interest and the whole brain regions in the preprocessed functional magnetic resonance imaging data.

[0092] The second matrix acquisition module 840 is used to obtain a functional connection weight matrix between the depression scale score data and the functional connection matrix, and the functional connection weight matrix is ​​used to characterize the degree of correlation between the functional connection matrix and the depression scale score data.

[0093] The target acquisition module 850 is used to map the functional connection weight matrix into the functional connection matrix to obtain an estimated therapeutic effect map of the whole brain area; cluster the voxels in the left DLPFC area in the estimated therapeutic effect map, and determine the cluster center as the TMS treatment target.

[0094] In a possible implementation, the region acquisition module 820 is specifically configured to: The coil position and coil posture of each subject during TMS treatment are obtained; based on the coil position and coil posture, electric field simulation is performed in the cerebral cortex of the preprocessed structural magnetic resonance imaging data to obtain a simulated electric field; the point with the maximum simulated electric field value in the simulated electric field is obtained as the TMS stimulation target; in the cerebral cortex, the region of interest corresponding to the TMS stimulation target of each subject is obtained to obtain a plurality of regions of interest.

[0095] In a possible implementation manner, the first matrix acquisition module 830 is specifically configured to: Extracting blood oxygen level dependent signals of several regions of interest, whole brain regions and whole brain voxels to obtain time series matrices of several regions of interest, whole brain regions and whole brain voxels; Obtain the Pearson correlation coefficient between the time series matrices of several regions of interest and the time series matrices of the whole brain region to obtain the first functional connectivity matrix; Obtain the Pearson correlation coefficient between the time series matrix of the whole brain area and the time series matrix of the whole brain voxels to obtain the second functional connectivity matrix; A functional connectivity matrix is ​​determined based on the first functional connectivity matrix and the second functional connectivity matrix.

[0096] In a possible implementation, the second matrix acquisition module 840 is used to acquire a functional connection weight matrix between depression rating scale score data and a functional connection matrix based on a partial least squares regression model.

[0097] In a possible implementation, the image acquisition module 810 is specifically configured to: Acquiring brain tissue image data from structural magnetic resonance image data; Acquire functional magnetic resonance imaging data after time correction and head motion correction to obtain whole-brain imaging data; The brain tissue image data and the whole brain region image data are registered to the same standard template to obtain the preprocessed structural magnetic resonance image data and the preprocessed functional magnetic resonance image data, respectively.

[0098] In a possible implementation, the image acquisition module 810 is further configured to perform smoothing filtering on the extracted blood oxygen level dependent signal through a bandpass filter.

[0099] Fig. 9 An example of a physical structure diagram of an electronic device is shown in FIG. Fig. 9As shown, the electronic device may include: a processor 910, a communication interface 920, a memory 930 and a communication bus 940, wherein the processor 910, the communication interface 920 and the memory 930 communicate with each other through the communication bus 940. The processor 910 may call the logic instructions in the memory 930 to execute a TMS target location method based on functional connection weights, the method comprising: Obtaining depression scale score data, structural magnetic resonance imaging data, and functional magnetic resonance imaging data of several subjects; preprocessing the structural magnetic resonance imaging data and the functional magnetic resonance imaging data to obtain preprocessed structural magnetic resonance imaging data and preprocessed functional magnetic resonance imaging data; In the preprocessed structural magnetic resonance imaging data, the region of interest corresponding to the TMS stimulation target of each subject is obtained to obtain a number of regions of interest; Obtaining functional connectivity matrices between several regions of interest and whole-brain brain regions in preprocessed functional magnetic resonance imaging data; Obtaining a functional connection weight matrix between the depression scale score data and the functional connection matrix, wherein the functional connection weight matrix is ​​used to characterize the degree of correlation between the functional connection matrix and the depression scale score data; The functional connection weight matrix was mapped to the functional connection matrix to obtain the estimated therapeutic effect map of the whole brain area; the voxels in the left DLPFC area in the estimated therapeutic effect map were clustered, and the cluster center was determined as the TMS treatment target.

[0100] In addition, the logic instructions in the above-mentioned memory 930 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0101] On the other hand, the present invention further provides a computer program product, the computer program product comprising a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer can execute a TMS target location method based on functional connection weights provided by the above methods, the method comprising: Obtaining depression scale score data, structural magnetic resonance imaging data, and functional magnetic resonance imaging data of several subjects; preprocessing the structural magnetic resonance imaging data and the functional magnetic resonance imaging data to obtain preprocessed structural magnetic resonance imaging data and preprocessed functional magnetic resonance imaging data; In the preprocessed structural magnetic resonance imaging data, the region of interest corresponding to the TMS stimulation target of each subject is obtained to obtain a number of regions of interest; Obtaining functional connectivity matrices between several regions of interest and whole-brain brain regions in preprocessed functional magnetic resonance imaging data; Obtaining a functional connection weight matrix between the depression scale score data and the functional connection matrix, wherein the functional connection weight matrix is ​​used to characterize the degree of correlation between the functional connection matrix and the depression scale score data; The functional connection weight matrix was mapped to the functional connection matrix to obtain the estimated therapeutic effect map of the whole brain area; the voxels in the left DLPFC area in the estimated therapeutic effect map were clustered, and the cluster center was determined as the TMS treatment target.

[0102] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to perform a TMS target location method based on functional connection weights provided by the above methods, the method comprising: Obtaining depression scale score data, structural magnetic resonance imaging data, and functional magnetic resonance imaging data of several subjects; preprocessing the structural magnetic resonance imaging data and the functional magnetic resonance imaging data to obtain preprocessed structural magnetic resonance imaging data and preprocessed functional magnetic resonance imaging data; In the preprocessed structural magnetic resonance imaging data, the region of interest corresponding to the TMS stimulation target of each subject is obtained to obtain a number of regions of interest; Obtaining functional connectivity matrices between several regions of interest and whole-brain brain regions in preprocessed functional magnetic resonance imaging data; Obtaining a functional connection weight matrix between the depression scale score data and the functional connection matrix, wherein the functional connection weight matrix is ​​used to characterize the degree of correlation between the functional connection matrix and the depression scale score data; The functional connection weight matrix was mapped to the functional connection matrix to obtain the estimated therapeutic effect map of the whole brain area; the voxels in the left DLPFC area in the estimated therapeutic effect map were clustered, and the cluster center was determined as the TMS treatment target.

[0103] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0104] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A TMS target localization method based on functional connection weights, characterized in that: include: Obtain depression scale score data, structural magnetic resonance imaging data, and functional magnetic resonance imaging data of several subjects; Preprocessing the structural magnetic resonance imaging data and the functional magnetic resonance imaging data to obtain preprocessed structural magnetic resonance imaging data and preprocessed functional magnetic resonance imaging data; In the preprocessed structural magnetic resonance imaging data, a region of interest corresponding to the TMS stimulation target of each subject is obtained to obtain a plurality of regions of interest; Acquire a functional connection matrix between the plurality of regions of interest and the whole brain regions in the preprocessed functional magnetic resonance imaging data; Obtaining a functional connection weight matrix between the depression rating scale score data and the functional connection matrix, wherein the functional connection weight matrix is ​​used to characterize the degree of correlation between the functional connection matrix and the depression rating scale score data; The functional connection weight matrix is ​​mapped to the functional connection matrix to obtain an estimated therapeutic effect map of the whole brain area; the voxels in the left DLPFC area in the estimated therapeutic effect map are clustered, and the cluster center is determined as the TMS treatment target.

2. The TMS target localization method based on functional connection weight according to claim 1, characterized in that: In the pre-processed structural magnetic resonance imaging data, a region of interest corresponding to the TMS stimulation target of each subject is obtained to obtain a plurality of regions of interest, including: Obtain the coil position and coil posture of each subject during TMS treatment; According to the coil position and the coil posture, electric field simulation is performed in the cerebral cortex of the preprocessed structural magnetic resonance image data to obtain a simulated electric field; Obtaining the point with the maximum simulated electric field value in the simulated electric field as the TMS stimulation target point; In the cerebral cortex, the region of interest corresponding to the TMS stimulation target of each subject is obtained to obtain a plurality of regions of interest.

3. The TMS target localization method based on functional connection weight according to claim 1, characterized in that: The step of obtaining a functional connection matrix between the plurality of regions of interest and the whole brain regions in the preprocessed functional magnetic resonance imaging data comprises: Extracting blood oxygen level dependent signals of the plurality of regions of interest, the plurality of whole brain regions and the plurality of whole brain voxels to obtain time series matrices of the plurality of regions of interest, the plurality of whole brain regions and the plurality of whole brain voxels; Obtaining the Pearson correlation coefficient between the time series matrices of the plurality of regions of interest and the time series matrices of the entire brain region to obtain a first functional connectivity matrix; Obtaining the Pearson correlation coefficient between the time series matrix of the whole brain region and the time series matrix of the whole brain voxels to obtain a second functional connectivity matrix; The functional connectivity matrix is ​​determined according to the first functional connectivity matrix and the second functional connectivity matrix.

4. The TMS target localization method based on functional connection weight according to claim 1, characterized in that: The step of obtaining a functional connection weight matrix between the depression rating scale score data and the functional connection matrix comprises: Based on the partial least squares regression model, a functional connection weight matrix between the depression rating scale score data and the functional connection matrix is ​​obtained.

5. The TMS target localization method based on functional connection weight according to claim 1, characterized in that: The preprocessing of the structural magnetic resonance imaging data and the functional magnetic resonance imaging data to obtain the preprocessed structural magnetic resonance imaging data and the preprocessed functional magnetic resonance imaging data includes: Acquiring brain tissue image data from the structural magnetic resonance image data; Acquiring the functional magnetic resonance imaging data after time correction and head motion correction to obtain corrected functional magnetic resonance imaging data; The brain tissue image data and the corrected functional magnetic resonance image data are registered on the same standard template to obtain the preprocessed structural magnetic resonance image data and the preprocessed functional magnetic resonance image data respectively.

6. The TMS target location method based on functional connection weight according to claim 3, characterized in that: Also includes: The extracted blood oxygen level dependent signal is smoothed by a bandpass filter.

7. A TMS target positioning system based on functional connection weights, characterized in that: include: An image acquisition module, used to acquire depression scale score data, structural magnetic resonance imaging data, and functional magnetic resonance imaging data of several subjects; Preprocessing the structural magnetic resonance imaging data and the functional magnetic resonance imaging data to obtain preprocessed structural magnetic resonance imaging data and preprocessed functional magnetic resonance imaging data; A region acquisition module, used to acquire a region of interest corresponding to a TMS stimulation target of each subject in the preprocessed structural magnetic resonance imaging data, and obtain a plurality of regions of interest; A first matrix acquisition module is used to acquire a functional connection matrix between the plurality of regions of interest and the whole brain regions in the preprocessed functional magnetic resonance imaging data; A second matrix acquisition module is used to acquire a functional connection weight matrix between the depression scale score data and the functional connection matrix, wherein the functional connection weight matrix is ​​used to characterize the degree of correlation between the functional connection matrix and the depression scale score data; The target acquisition module is used to map the functional connection weight matrix into the functional connection matrix to obtain an estimated therapeutic effect map of the whole brain area; cluster the voxels in the left DLPFC area in the estimated therapeutic effect map, and determine the cluster center as the TMS treatment target.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the TMS target localization method based on functional connection weights as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the TMS target localization method based on functional connection weights as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the TMS target localization method based on functional connection weights as described in any one of claims 1 to 6 is implemented.

Images